Unraveling the functional diversity of B cells in health and disease

B细胞在传染病、自身免疫和过敏中的作用至关重要。现代测序 单细胞RNA测序(ScRNAseq)和空间转录组学等技术已经 作为研究个体B细胞转录状态的强大技术出现在各种 在生物学背景下。这些技术会产生海量的复杂数据 有必要使用强大、复杂的计算方法。对这些数据的分析是 受制于数据中嵌入的大量技术和生物偏见。在scRNAseq中，for 例如，沿着某些发育轨迹对细胞的非均匀捕获，以及 多个并发转录程序的表达，对目前的单细胞构成了挑战 聚类法和轨迹推理法。当研究B细胞时，这些偏见会加剧 具有复杂动力学的隔室，如在淋巴组织中发现的隔室。要解决这些问题 问题，我们提出了一种新的算法工具箱，用于建模B细胞活动，它结合了先前的， 通过计算算法设计验证生物学知识。在目标1中，我们开发工具来 阐明暂时性B细胞的发育过程。在目标2中，我们开发了工具来阐明B细胞 空间转录程序。在目标3中，将我们的工具应用于各种重要的临床场景， 例如绘制高亲和力抗体的免疫关联图，以及表征 在IBD中观察到异质性。总体而言，我们的研究将为 分析scRNAseq和空间转录数据中的免疫信号，以及显示 结合先验知识极大地提高了计算算法揭示 免疫系统的全方位变化，以应对疫苗接种、感染和 免疫介导的疾病。

The role of B cells in infectious disease, autoimmunity, and allergy is critical. Modern sequencing technologies, such as single-cell RNA sequencing (scRNAseq) and spatial transcriptomics, have emerged as powerful techniques for studying the transcriptional states of individual B cells in a variety of biological contexts. These technologies generate massive amounts of complex data that necessitate use of powerful, sophisticated computational methods. The analysis of such data is hampered by numerous technical and biological biases embedded in the data. In scRNAseq, for example, the non-uniform capture of cells along some developmental trajectory, as well as the expression of multiple concurrent transcriptional programs, pose a challenge to current single cell clustering and trajectory inference methods. These biases are exacerbated when studying B cell compartments with complex dynamics, such as those found in lymphoid tissues. To address these issues, we propose a novel toolbox of algorithms for modeling B cell activity that combines prior, validated biological knowledge with computational algorithm design. In Aim 1, we develop tools to elucidate temporal B cell developmental processes. And in Aim 2, we develop tools to elucidate B cell spatial transcriptional programs. In Aim 3, apply our tools to a variety of important clinical scenarios, such as mapping the immune correlates of higher affinity antibodies and characterizing the heterogeneity observed in IBD. Overall, our research will create much-needed computational tools for analyzing immune signals in scRNAseq and spatial transcriptomics data, as well as show that incorporating prior knowledge greatly improves the ability of computational algorithms to reveal the full spectrum of immune system changes that occur in response to vaccination, infection, and immune-mediated diseases.